FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3177684257> ?p ?o ?g. }

• W3177684257 abstract "Scene text recognition (STR) task has a common practice: All state-of-the-art STR models are trained on large synthetic data. In contrast to this practice, training STR models only on fewer real labels (STR with fewer labels) is important when we have to train STR models without synthetic data: for handwritten or artistic texts that are difficult to generate synthetically and for languages other than English for which we do not always have synthetic data. However, there has been implicit common knowledge that training STR models on real data is nearly impossible because real data is insufficient. We consider that this common knowledge has obstructed the study of STR with fewer labels. In this work, we would like to reactivate STR with fewer labels by disproving the common knowledge. We consolidate recently accumulated public real data and show that we can train STR models satisfactorily only with real labeled data. Subsequently, we find simple data augmentation to fully exploit real data. Furthermore, we improve the models by collecting unlabeled data and introducing semi- and self-supervised methods. As a result, we obtain a competitive model to state-of-the-art methods. To the best of our knowledge, this is the first study that 1) shows sufficient performance by only using real labels and 2) introduces semi- and self-supervised methods into STR with fewer labels. Our code and data are available: https://github.com/ku21fan/STR-Fewer-Labels." @default.
• W3177684257 created "2021-07-19" @default.
• W3177684257 creator A5023905620 @default.
• W3177684257 creator A5065773265 @default.
• W3177684257 creator A5069982192 @default.
• W3177684257 date "2021-06-01" @default.
• W3177684257 modified "2023-10-16" @default.
• W3177684257 title "What If We Only Use Real Datasets for Scene Text Recognition? Toward Scene Text Recognition With Fewer Labels" @default.
• W3177684257 cites W1971822075 @default.
• W3177684257 cites W1981283549 @default.
• W3177684257 cites W2008806374 @default.
• W3177684257 cites W2122585011 @default.
• W3177684257 cites W2127141656 @default.
• W3177684257 cites W2128409098 @default.
• W3177684257 cites W2140132917 @default.
• W3177684257 cites W2144554289 @default.
• W3177684257 cites W2146835493 @default.
• W3177684257 cites W2194775991 @default.
• W3177684257 cites W2343052201 @default.
• W3177684257 cites W2785383245 @default.
• W3177684257 cites W2798991696 @default.
• W3177684257 cites W2914492226 @default.
• W3177684257 cites W2963587345 @default.
• W3177684257 cites W2963712589 @default.
• W3177684257 cites W2964065044 @default.
• W3177684257 cites W2965066169 @default.
• W3177684257 cites W2965463054 @default.
• W3177684257 cites W2970910956 @default.
• W3177684257 cites W2979382951 @default.
• W3177684257 cites W2988326850 @default.
• W3177684257 cites W2997749585 @default.
• W3177684257 cites W3003218881 @default.
• W3177684257 cites W3003711889 @default.
• W3177684257 cites W3003868038 @default.
• W3177684257 cites W3003921261 @default.
• W3177684257 cites W3005400651 @default.
• W3177684257 cites W3005436539 @default.
• W3177684257 cites W3035449864 @default.
• W3177684257 cites W3103667010 @default.
• W3177684257 cites W3110267192 @default.
• W3177684257 doi "https://doi.org/10.1109/cvpr46437.2021.00313" @default.
• W3177684257 hasPublicationYear "2021" @default.
• W3177684257 type Work @default.
• W3177684257 sameAs 3177684257 @default.
• W3177684257 citedByCount "29" @default.
• W3177684257 countsByYear W31776842572021 @default.
• W3177684257 countsByYear W31776842572022 @default.
• W3177684257 countsByYear W31776842572023 @default.
• W3177684257 crossrefType "proceedings-article" @default.
• W3177684257 hasAuthorship W3177684257A5023905620 @default.
• W3177684257 hasAuthorship W3177684257A5065773265 @default.
• W3177684257 hasAuthorship W3177684257A5069982192 @default.
• W3177684257 hasBestOaLocation W31776842572 @default.
• W3177684257 hasConcept C119857082 @default.
• W3177684257 hasConcept C154945302 @default.
• W3177684257 hasConcept C160920958 @default.
• W3177684257 hasConcept C162324750 @default.
• W3177684257 hasConcept C165696696 @default.
• W3177684257 hasConcept C177264268 @default.
• W3177684257 hasConcept C187736073 @default.
• W3177684257 hasConcept C199360897 @default.
• W3177684257 hasConcept C204321447 @default.
• W3177684257 hasConcept C2776145971 @default.
• W3177684257 hasConcept C2776760102 @default.
• W3177684257 hasConcept C2780451532 @default.
• W3177684257 hasConcept C38652104 @default.
• W3177684257 hasConcept C41008148 @default.
• W3177684257 hasConcept C51632099 @default.
• W3177684257 hasConceptScore W3177684257C119857082 @default.
• W3177684257 hasConceptScore W3177684257C154945302 @default.
• W3177684257 hasConceptScore W3177684257C160920958 @default.
• W3177684257 hasConceptScore W3177684257C162324750 @default.
• W3177684257 hasConceptScore W3177684257C165696696 @default.
• W3177684257 hasConceptScore W3177684257C177264268 @default.
• W3177684257 hasConceptScore W3177684257C187736073 @default.
• W3177684257 hasConceptScore W3177684257C199360897 @default.
• W3177684257 hasConceptScore W3177684257C204321447 @default.
• W3177684257 hasConceptScore W3177684257C2776145971 @default.
• W3177684257 hasConceptScore W3177684257C2776760102 @default.
• W3177684257 hasConceptScore W3177684257C2780451532 @default.
• W3177684257 hasConceptScore W3177684257C38652104 @default.
• W3177684257 hasConceptScore W3177684257C41008148 @default.
• W3177684257 hasConceptScore W3177684257C51632099 @default.
• W3177684257 hasLocation W31776842571 @default.
• W3177684257 hasLocation W31776842572 @default.
• W3177684257 hasOpenAccess W3177684257 @default.
• W3177684257 hasPrimaryLocation W31776842571 @default.
• W3177684257 hasRelatedWork W2033364610 @default.
• W3177684257 hasRelatedWork W2130553454 @default.
• W3177684257 hasRelatedWork W2797776314 @default.
• W3177684257 hasRelatedWork W2949671220 @default.
• W3177684257 hasRelatedWork W3008176773 @default.
• W3177684257 hasRelatedWork W3021676282 @default.
• W3177684257 hasRelatedWork W3022007134 @default.
• W3177684257 hasRelatedWork W3175965105 @default.
• W3177684257 hasRelatedWork W4312659495 @default.
• W3177684257 hasRelatedWork W4317548404 @default.
• W3177684257 isParatext "false" @default.
• W3177684257 isRetracted "false" @default.
• W3177684257 magId "3177684257" @default.

• next